cs.LGAug 14, 2026

M-LINKX: Multiview Graph Learning for Brain Cognitive Disease Detection

Authors: An Phan, Yufei Jin, Xingquan Zhu

Organizations: Dept. of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL 33431, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Electroencephalogram (EEG) is a non-invasive and relatively low-cost procedure that measures brain electricity for the detection of cognitive diseases. EEG-based classification of dementia-related conditions, including Alzheimer's disease (AD), mild cognitive impairment (MCI), and frontotemporal dementia (FTD), remains challenging because EEG signals are noisy, non-stationary, and vary across subjects. Segment-based learning provides a practical way to model long EEG recordings by converting them into fixed-length inputs. For each segment, discriminative information may be explored by using signals within each channel (i.e. electrode), as well as interactions between EEG channels. In this paper, we propose M-LINKX, a multi-view graph learning framework for EEG-based dementia classification. For each segment, we extract channel-level node features and construct multiple functional-connectivity (FC) graph views, where each view is defined by a specific combination of connectivity metric, frequency band, and topology filter, respectively. Instead of relying on message passing over the constructed graphs, M-LINKX follows a simple design in modeling node features and adjacency-based connectivity representations. The graph-view representations are fused using global trainable view weights, and subject-level prediction is obtained by averaging segment-level probabilities. Experiments on two three-class EEG datasets with different diagnostic groups, CAUEEG (HC/MCI/Dementia) and AHEAP (HC/AD/FTD), show that M-LINKX achieves the best subject-level performance under the main experimental settings. Our study suggests that multi-view functional connectivity can improve EEG-based dementia classification when integrated with an appropriate graph-learning architecture. Code and data are available at https://github.com/anphantt/MLINKX.

Figures & tables

Fig. 1 : Overview of the proposed M-LINKX framework. For readability, block ① should be read from top to bottom, whereas blocks ② and ③ should be read from left to right. Block ① shows the subject-level classification pipeline: each EEG recording is divided into fixed-length overlapping segments, segment-level probabilities are predicted, and subject-level prediction is obtained by averaging probabilities across segments. Block ② shows feature extraction and graph construction for a segment, where “segment 2” is represented by channel-level node features and multiple FC graph views from different connectivity metrics, frequency bands, and topology filters. Block ③ shows the M-LINKX encoder, which learns node-feature and graph-view embeddings, fuses graph-view embeddings using global trainable view weights, and produces segment-level predictions.
Fig. 2 : Topology-specific graph representations of alpha-band coherence from three consecutive EEG segments of the same subject recording. Nodes denote EEG channels arranged by approximate 10–20 scalp locations, and weighted edges denote retained coherence connections. Rows show the four topology filters: complete , domain , top- kk , and hybrid ; columns show consecutive EEG segments. Edge thickness and color indicate coherence weight, with thicker and darker blue edges representing larger weights and thinner gray edges representing smaller weights.
PropertyAHEAPCAUEEG
# of Subjects36 AD, 23 FTD, 29 HC291 Dementia, 395 MCI, 436 HC
# of Channels1919
ConditionEyes closedEyes closed and eyes open
Sampling Rate500 Hz200 Hz
Avg. Duration13.21 ±\pm 2.32 min12.80 ±\pm 2.65 min
TABLE I : Summary of the EEG datasets used in this study.
    CAUEEG (10-second window)AHEAP (4-second window)
    35 HC, 33 MCI, 22 Dementia29 HC, 23 FTD, 36 AD
ModelRawEEGNode featuresFC inputBalanced Acc.Macro-F1Balanced Acc.Macro-F1
1D ResNet✓  0.6146 ±\pm 0.01350.6224 ±\pm 0.01410.5221 ±\pm 0.04320.4815 ±\pm 0.0511
CNN-LSTM✓  0.5700 ±\pm 0.01330.5585 ±\pm 0.00490.5697 ±\pm 0.05270.5494 ±\pm 0.0438
MLP-node ✓ 0.5949 ±\pm 0.02330.5961 ±\pm 0.01990.5740 ±\pm 0.04770.5550 ±\pm 0.0446
Ensemble✓✓ 0.6084 ±\pm 0.02850.6048 ±\pm 0.03030.5791 ±\pm 0.02580.5528 ±\pm 0.0273
TABLE II : Subject-level performance of M-LINKX and the baseline models, reported as mean ±\pm standard deviation over three random seeds. Best and second-best results are shown in bold and underlined italics , respectively. For methods involving different Functional connectivity (“FC Input”), V\mathcal{V} denotes all four candidate graph views and v⋆v^{\star} denotes the best-performing view among the four candidate graph views V\mathcal{V} .
Fig. 3 : Subject-level confusion matrices of M-LINKX on CAUEEG (10 ss ) and AHEAP (4 ss ). Cells show average subject counts and row-normalized percentages over three random seeds. For AHEAP, confusion matrices are first summed across folds within each seed.
Fig. 4 : Subject-level balanced accuracy comparisons between M-LINKX and baseline models w.r.t. different segment durations on CAUEEG and AHEAP. yy -axis denotes mean balanced accuracies over three random seeds, and error bars denote standard deviation.
Fig. 5 : M-LINKX trainable graph-view weights across segment durations. Bars denote the mean weight over three random seeds, and error bars denote standard deviation.
More in this topic

Electroencephalography

Electroencephalography (EEG) studies brain-generated electrical activity to understand cognition and support diagnosis, monitoring, and brain-computer interfaces.

See all 180 papers

Explore similar work

CardsList
Mar 3, 2026

EEG-SeeGraph: Interpreting functional connectivity disruptions in dementias via sparse-explanatory dynamic EEG-graph learning

Robust and interpretable dementia diagnosis from noisy, non-stationary electroencephalography (EEG) is clinically essential yet remains challenging. To this end, we propose SeeGraph, a Sparse-Explanatory dynamic EEG-graph network that models time-evolving functional connectivity and employs a node-guided sparse edge mask to reveal the connections that drive diagnostic decisions, while remaining robust to noise and cross-site variability. SeeGraph comprises four components: (1) a dual-trajectory temporal encoder that models dynamic EEG with two streams, where node signals capture regional oscillations and edge signals capture interregional coupling; (2) a topology-aware positional encoder that derives graph-spectral Laplacian coordinates from the fused connectivity and augments node embeddings; (3) a node-guided sparse explanatory edge mask that gates the connectivity into a compact subgraph; and (4) a gated graph predictor that operates on the sparsified graph. The framework is trained using cross-entropy loss together with a sparsity regularizer on the mask, yielding noise-robust and interpretable diagnoses. The effectiveness of SeeGraph is validated on public and in-house EEG cohorts, including patients with neurodegenerative dementias and healthy controls, under both raw and noise-perturbed conditions. Its sparse, node-guided explanations highlight disease-relevant connections and align with established clinical findings on functional connectivity alterations, thereby offering transparent cues for neurological evaluation.

Fengcheng Wu, Zhenxi Song, Guoyang Xu +4

Harbin Institute of Technology, Shenzhen, China · Institute of Neurological Diseases, Shenzhen Bay Laboratory, Shenzhen · Institute of Neurological Diseases, Shenzhen Bay Laboratory, Shenzhen, China

ElectroencephalographyDynamic Graph LearningFunctional ConnectivityGNN Explainability
AskPDF
Apr 27, 2026

Task-guided Spatiotemporal Network with Diffusion Augmentation for EEG-based Dementia Diagnosis and MMSE Prediction

Patients with dementia typically exhibit cognitive impairment, which is routinely assessed using the Mini-Mental State Examination (MMSE). Concurrently, their underlying neurophysiological abnormalities are reflected in Electroencephalography (EEG), providing a basis for joint modeling. However, traditional multi-task approaches suffer from feature entanglement, which leads to inter-task interference when handling heterogeneous objectives.To address this challenge, we propose a task-guided spatiotemporal network (TGSN) with diffusion augmentation for EEG-based dementia diagnosis and MMSE prediction. Specifically, TGSN integrates a multi-band feature fusion module to capture complementary spectral information from EEG. Meanwhile, a pre-trained data augmentation module utilizing a diffusion process is introduced toincrease sample diversity. To model the complex spatiotemporal patterns of EEG, we propose a gated spatiotemporal attention module that captures long-range spatial dependencies and temporal dynamics. Moreover, we design a task-guided query module to achieve task-specific feature extraction, thereby mitigating task interference. The effectiveness of TGSN is evaluated on the XY02 dataset. Experimental results demonstrate that the proposed network outperforms several state-of-the-art methods, achieving classification accuracies of 97.78% for Alzheimer's Disease (AD)/Frontotemporal Dementia (FTD) and 83.93% for AD/FTD/Vascular Cognitive Impairment (VCI), which exceed the best baselines by 16.39% and 8.28%, respectively. In parallel, it reduces the RMSE for MMSE prediction to 1.93 and 2.38, achieving significant error reductions of 1.44 and 1.43 compared to the best baselines. Additionally, validation on the DS004504 dataset demonstrates strong cross-dataset generalization...

Xiaoyu Zheng, Xu Tian, Bin Jiao +4

Xinjiang Engineering Research Center of Big Data and Intelligent Software, School of Software, Xinjiang University, Urumqi 830091, China. · Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha 410083, China. · Department of Neurology, Xiangya Hospital, Central South University, Changsha 410083, China. +1

Alzheimer's DiseaseElectroencephalographyEEG DecodingMulti-Task LearningAlzheimer's Disease Classification
AskPDF
Oct 13, 2025

Variational Mixture of Graph Neural Experts for Alzheimer's Disease Recognition across Frequency Bands in EEG Brain Networks

Dementia disorders such as Alzheimer's disease (AD) and frontotemporal dementia (FTD) exhibit overlapping electrophysiological signatures in EEG that challenge accurate diagnosis. Existing EEG-based methods are limited by full-band frequency analysis, which hinders the precise differentiation of dementia subtypes and severity stages. To address this limitation, we propose a Variational Mixture of Graph Neural Experts (VMoGE) framework that integrates multi-band EEG analysis with variational graph neural networks and a mixture-of-experts architecture. Each expert specializes in a specific EEG frequency band and models brain connectivity using a Gaussian Markov random field prior, while a variational gating mechanism adaptively integrates the expert outputs. This design enables the model to learn frequency-specific brain network representations while modeling latent uncertainty through variational inference. Experimental results on two EEG dementia datasets show that VMoGE achieves strong performance, with an AUC of 0.89 for HC vs. AD classification in the main comparison and competitive results across dementia subtyping and CDR staging tasks. Clinically, VMoGE offers three key translational values: the expert gating weights correlate with MMSE scores and CDR severity; slow-wave δδ- and θθ-band contributions are associated with AD-related EEG slowing and disease progression; and spatially localized activation maps reveal posterior θθ- and αα-band alterations and region-specific ββ-band changes, providing neurophysiologically interpretable markers aligned with known AD neuropathology.

Jun-En Ding, Anna Zilverstand, Shihao Yang +2

Department of Industrial and Systems Engineering, Rutgers University, Piscataway, NJ, USA · Department of Psychiatry and Behavioral Sciences, University of Minnesota, Minneapolis, MN, USA · Institute of Brain Science, College of Medicine, National Yang-Ming Chiao Tung University, Taipei City, Taiwan

Alzheimer's DiseaseElectroencephalographyGraph Neural NetworksFrequency-Domain Feature LearningAlzheimer's Disease ClassificationMixture-of-Experts Models
AskPDF